The original 3 channel stereo had it. You can matrix a centre channel from the stereo and often get similar results. Narrows the soundstage over regular 2 channel of course, but combines a bit of mono and a bit of stereo. Worth trying out. I used to run the three channel ambiophonic setup with twin rear speakers, and it really can work very well.
The horizontal array was the beginning of Ted's work on stable left-right soundfields from a stereo system, wherever you stood in the room. There was a cut down version which made use of reflectors. The drivers were back to back in the centre, facing left and right, you then put a hard reflector in the positions you'd normally have your speakers. The sound from the units in the middle stablised the image and the reflectors gave you left and right.
I tried this some years ago with a highly advanced set up, viz a couple of cheap fullrangers and a couple of formica-topped coffee tables tipped on end. It actually worked. I ought to have another go now I have something more high end in the speaker line (although I no longer have the formica tables).
The fun thing today, of course, is it gives you a movie centre speaker without the extra fuss and the reflecting surfaces could be made to look like a pair of open baffles. Confound all your friends and claim the baffles operate via wi-fi!
it was all part of the same article, published in Wireless World in Feb 1971. I've been meaning to copy it for a while so if I get the chance to scan it in the next few days, I can email you a copy, Brian. If you would like to go to my website you'll find my email address there, if you'd like to get in touch.
I tried this some years ago with a highly advanced set up, viz a couple of cheap fullrangers and a couple of formica-topped coffee tables tipped on end. It actually worked. I ought to have another go now I have something more high end in the speaker line (although I no longer have the formica tables).
The fun thing today, of course, is it gives you a movie centre speaker without the extra fuss and the reflecting surfaces could be made to look like a pair of open baffles. Confound all your friends and claim the baffles operate via wi-fi!
it was all part of the same article, published in Wireless World in Feb 1971. I've been meaning to copy it for a while so if I get the chance to scan it in the next few days, I can email you a copy, Brian. If you would like to go to my website you'll find my email address there, if you'd like to get in touch.
cabinet stuffing
As I ponder the final design wonder what you've found with stuffing materials. In our hardgoods neighborhood here in Mongkok one can buy the typical acoustic foam, fibreglass ceiling squares and some 2mm thick sheet good, metal impregnated. So far haven't found anyone that sells rockwool.
In our shallow speaker cabs I was getting some zing from female vocals especially, certain notes only. I suppose the distance is quite short from back of driver to the back wall so the reflection is coming back at near full amplitude. A mic through the port into the cabinet showed a spike at 2k. That was with the box stuffed with acoustic foam. I lined the box with the 2mm sheet material and then put in a little foam to hold it in place. That pretty well killed the zing and measurement showed the spike was gone. Interesting that sheet stuff gets a bit warm when hit with high levels of pink noise.
The sheet material comes with an attenuation chart at frequency/db and claims >20db attenuation for 1k up.
Are there any other materials that could give better attenuation? Has anyone ever measure frq vs attenuation for different material types? Would be a valuable resource to be able to select the material that gives the attenuation at the problem frequency.
Also noticed that stuffing the cabinet full of foam seems to slow down the air movement internally and kill the dynamics. But the sheet material seems to 'trap' the frequencies and speaker sounds more lively. Do stuffing materials generally excel at one or the other?
Or are my subjective observations leading me astray?
Brian
As I ponder the final design wonder what you've found with stuffing materials. In our hardgoods neighborhood here in Mongkok one can buy the typical acoustic foam, fibreglass ceiling squares and some 2mm thick sheet good, metal impregnated. So far haven't found anyone that sells rockwool.
In our shallow speaker cabs I was getting some zing from female vocals especially, certain notes only. I suppose the distance is quite short from back of driver to the back wall so the reflection is coming back at near full amplitude. A mic through the port into the cabinet showed a spike at 2k. That was with the box stuffed with acoustic foam. I lined the box with the 2mm sheet material and then put in a little foam to hold it in place. That pretty well killed the zing and measurement showed the spike was gone. Interesting that sheet stuff gets a bit warm when hit with high levels of pink noise.
The sheet material comes with an attenuation chart at frequency/db and claims >20db attenuation for 1k up.
An externally hosted image should be here but it was not working when we last tested it.
Are there any other materials that could give better attenuation? Has anyone ever measure frq vs attenuation for different material types? Would be a valuable resource to be able to select the material that gives the attenuation at the problem frequency.
Also noticed that stuffing the cabinet full of foam seems to slow down the air movement internally and kill the dynamics. But the sheet material seems to 'trap' the frequencies and speaker sounds more lively. Do stuffing materials generally excel at one or the other?
Or are my subjective observations leading me astray?
Brian
That is one way...they let loose little (software) robots or spiders that crawl all over the web looking for something like... this.looks@likeanemailaddress.com and then when the robots find stuff they send it back to their evil masters....
Colin has kindly sent a copy of the Wireless World article on the horizontal line array. It is fascinating stuff although it doesn't give enough details to construct the time delay network.
Ted's article give a new reason (for me) to appreciate the imaging qualities of a speaker - it's directly related to our perception of resolution and detail. When the brain receives information binaurally, time and space relationships are preserved and the dynamics of individual threads of sound, both macro and micro, are more discernable.
An example cited: trying to parse words spoken in a noisy environment while listening with one ear as compared listening to the same voice with both ears. It is the same with individual instrumental sounds standing out from the others when there is good binaural information retrieval.
Brian
Ted's article give a new reason (for me) to appreciate the imaging qualities of a speaker - it's directly related to our perception of resolution and detail. When the brain receives information binaurally, time and space relationships are preserved and the dynamics of individual threads of sound, both macro and micro, are more discernable.
An example cited: trying to parse words spoken in a noisy environment while listening with one ear as compared listening to the same voice with both ears. It is the same with individual instrumental sounds standing out from the others when there is good binaural information retrieval.
Brian
Re: cabinet stuffing
Your observations are correct.. "Stuffing" in any form provides a degree of acoustic resistance.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=75940&pagenumber=2
Try this material in a latex primer:
http://www.hytechsales.com/insulating_paint_additives.html
Multiple layers on every interior surface and on the driver frame.
................................
With respect to various array configurations..
Most image localization is a function of spl from around 2-11 kHz - particularly latteral localization. (..and phase from 80-400 Hz can alter image position in the depth field.)
The Jordan drivers with their large sd's (..relative to a tweeter in this range), will exhibit even greater sp-loss off-axis.
If you "toe-in" the drivers to cross in "front" of the listener then imaging becomes more centered between the speakers. If the "toe-in" has the drivers crossing "behind" the listener then imaging extends further latterally/horizontally.
If you are willing to make some compromises on loudspeaker/listener placement, (compromises are to decor - not audio), then the best of both worlds, (and more), can be achieved by moving the speakers closer together with "toe-in" in the "behind" position, AND moving the speakers closer to the listener near the center of the room.
If you are concerned about comb filtering then you can always utilize a 5 driver bessel array (..with a loss in spl below 2kHz relative to 5 driver "normal" array).
Here is the Audience Au24 Array (..though it uses bandor 2 inch drivers instead of the Jordans):
http://www.stereotimes.com/CES_Au24.htm
bcherry said:
Your observations are correct.. "Stuffing" in any form provides a degree of acoustic resistance.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=75940&pagenumber=2
Try this material in a latex primer:
http://www.hytechsales.com/insulating_paint_additives.html
Multiple layers on every interior surface and on the driver frame.
................................
With respect to various array configurations..
Most image localization is a function of spl from around 2-11 kHz - particularly latteral localization. (..and phase from 80-400 Hz can alter image position in the depth field.)
The Jordan drivers with their large sd's (..relative to a tweeter in this range), will exhibit even greater sp-loss off-axis.
If you "toe-in" the drivers to cross in "front" of the listener then imaging becomes more centered between the speakers. If the "toe-in" has the drivers crossing "behind" the listener then imaging extends further latterally/horizontally.
If you are willing to make some compromises on loudspeaker/listener placement, (compromises are to decor - not audio), then the best of both worlds, (and more), can be achieved by moving the speakers closer together with "toe-in" in the "behind" position, AND moving the speakers closer to the listener near the center of the room.
If you are concerned about comb filtering then you can always utilize a 5 driver bessel array (..with a loss in spl below 2kHz relative to 5 driver "normal" array).
Here is the Audience Au24 Array (..though it uses bandor 2 inch drivers instead of the Jordans):
http://www.stereotimes.com/CES_Au24.htm
Feb 1971. I have a PDF if you'd like to email me.
ScottG - Ted's drivers are specifically designed to cross well in front of the listener, he's worked the cones radiation pattern to be narrow at HF. From a brief play with one of his array systems, I can vouch for his claims. In particular, if you toe out the enclosures to a more traditional 45 degrees, the image starts to collapse back to the enclosures. Toed in 60 degrees so they cross in front, the image spreads between the enclosures and is remarkably steady, irrespective of how far left or right you move.
I suspect the narrow radiation pattern of the Jordans (and possibly the Bandors) may negate some of the concerns about HF combing etc.
ScottG - Ted's drivers are specifically designed to cross well in front of the listener, he's worked the cones radiation pattern to be narrow at HF. From a brief play with one of his array systems, I can vouch for his claims. In particular, if you toe out the enclosures to a more traditional 45 degrees, the image starts to collapse back to the enclosures. Toed in 60 degrees so they cross in front, the image spreads between the enclosures and is remarkably steady, irrespective of how far left or right you move.
I suspect the narrow radiation pattern of the Jordans (and possibly the Bandors) may negate some of the concerns about HF combing etc.
ScottG
Great stuff. I'll try it.
Only concerned when I can hear it but so far not a trace with the 4driver array at any listening distance. The linked array is interesting. Has anyone heard it and any combing noticed? If they can do it...
soongsc
Colin has referenced it. If you do figure out the time delay network hope you share!
Colin
I suspect this too. Would it be possible to get a comment from Ted on the subject? If his driver is really breaking or even skirting the 'rule' against using full range drivers in a line array...
Brian
Great stuff. I'll try it.
Only concerned when I can hear it but so far not a trace with the 4driver array at any listening distance. The linked array is interesting. Has anyone heard it and any combing noticed? If they can do it...
soongsc
Colin has referenced it. If you do figure out the time delay network hope you share!
Colin
I suspect this too. Would it be possible to get a comment from Ted on the subject? If his driver is really breaking or even skirting the 'rule' against using full range drivers in a line array...
Brian
Colin said:
I understand what he is doing - in particular the driver's rise in response in the "imaging" region I mentioned works well. - i.e. it can allow for an off-axis listener position that is relativly flat while providing a more accentuated 0 degree axis that when rotated in-ward OR out-ward (i.e. toe-in or out), will enhance the particular imaging character as I noted before.
No doubt the the more traditional 45 degree angle you rotated to was coming close to positioning the 0 degree axis ON-axis (..that is the listener's axis). This is a BAD idea with driver that has a rising response in this region - and it subjectivly does what you mentioned.
Furthermore, I'd bet that the speakers were spread quite far apart - and some "center fill" was lost when compared to position/rotation before.
Trust me on this one - If you are really interested in improving sound quality (and don't have dometic limitations with decor), then try-out the loudspeaker/listener position I mentioned. The only "loss" to this configuration is the monophonic emphasis when moving around.
....
The narrow radiation pattern will not subjectivly alter comb-filtering effects - in this case it is solely that Ted has chosen a driver size that will minimize the effects to the point where comb-filtering really isn't objectionable (..and might even be subjectivly preferable to some listeners).
bcherry said:
I haven't heard it but I do think that "Kuze" PE project Scott linked to will give you a good idea of the eq. you will need to flatten the response in the midrange. Of course it also shows the comb filtering - but I doubt it will do anything more than subjectivly "fuzz" the "outline" of an image compared to a single driver or bessel array.
I chatted to Ted earlier today and took the opportunity of firing some of this at him. He said that he felt his horizontal array had been superceded by his current linear arrays, in as much as the linear array was cheaper and simpler to implement. His primary concern with both these has been stereo image stability. The horizontal system was also developed at a time when he was using a 4" cone full range unit (presumably the original Jordan Watts driver).
He also mentioned that the 4-unit array does the job perfectly well and the only real benefit from going to a longer array would be higher power handling for larger rooms, which seems to be what Brian is aiming at.
Scott - Thanks for the extra info. I don't have an array system at present but use the JX92, which has a rising response to mimic the effect of the array. I'll give the toe-out a try over the weekend. I've a feeling I've tried it before but I'll report back.
If I can find some suitable surfaces, I'll also have a go at the reflector idea mentioned in Ted's article.
There's a bit more info on the array on Ted's website. There was also more about it in an article he wrote for HiFi News in the 1980s, which described his original Jordan 50mm, how it was designed and how the linear array came to be developed. I probably have that somewhere so I'll see if I can dig it out (but no promises).
He also mentioned that the 4-unit array does the job perfectly well and the only real benefit from going to a longer array would be higher power handling for larger rooms, which seems to be what Brian is aiming at.
Scott - Thanks for the extra info. I don't have an array system at present but use the JX92, which has a rising response to mimic the effect of the array. I'll give the toe-out a try over the weekend. I've a feeling I've tried it before but I'll report back.
If I can find some suitable surfaces, I'll also have a go at the reflector idea mentioned in Ted's article.
There's a bit more info on the array on Ted's website. There was also more about it in an article he wrote for HiFi News in the 1980s, which described his original Jordan 50mm, how it was designed and how the linear array came to be developed. I probably have that somewhere so I'll see if I can dig it out (but no promises).
Colin said:
Concerning the horizonal arrays, after reading Ted's article, I think if one also reads an article BINAURAL RECORDINGS AND LOUDSPEAKERS by J. H. Buijs in Wireless World November 1982, one can probably understand the general concept of what's involved. This article is only for two speakers, but if you expand the equation for multiple speaker locations, you get a line array. I think it's worth a try.
One thing to consider when you use reflective surfaces is that once reflected, the dispersion of high frequencies become wider as appossed to the original beaming of the driver itself, so you will need a compensation circuit to compensate for expected high frequency roll-off.
Colin said:
You are correct - you don't need an array to try this.
However, remember that you want a reasnably flat response at your position when considering listening distance and the amount of "toe-out". The closer you are to the speakers, the more "toe-out" you can have and still maintain a flat response (..and this is one area where the line will produce a difference - i.e. you can listen further from the speakers and still maintain a flat response because of the -3db/meter linesource effect at higher freq.s vs. -6 db/meter effect for a pointsource).
In conjunction with all that you have center fill vs. increased width to contend with when moving the speakers further apart.
In otherwords it will take a fair bit of experimentation.
I'd suggest a meter and a half distance between you and the speakers with the driver's "toe-out" as far as possible, (aimed forward (no "toe-in") and even slightly beyond that), for a starting point. (i.e. like somewhat like a studio monitor set-up.) Moving youself forward to the speakers will allow moving the speakers futher apart.
(..note that one thing you should notice is that imaging, with certain recordings, will extend left of the left speaker and right of the right speaker.)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.